Markov processes and linguistics

Moishe Halibard, Ido Kanter

Research output: Contribution to journalArticlepeer-review

2 Scopus citations


The Markov Matrix Model of languages, created to explain patterns observed in sequences of data, is briefly reviewed. It is then extended to multiple-stage Markov processes to include the history of the process, to Markov processes with higher connectivities, and to Markov processes with varying transition probabilities. It is found that the simple restricted model produces distributions with all the features of these extended models. Correlations in binary sequences produced by the Markov chain are reviewed and compared to correlations in a related, more restricted chain, which is open to analytical investigation. The analysis on this chain is presented, and conclusions are drawn which may also shed light on the more general case.

Original languageEnglish
Pages (from-to)525-535
Number of pages11
JournalPhysica A: Statistical Mechanics and its Applications
Issue number1-4
StatePublished - 2 Jan 1998


Dive into the research topics of 'Markov processes and linguistics'. Together they form a unique fingerprint.

Cite this